| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Systemics and Phytogeography |
2Departamento de Biología Vegetal II, Facultad de Farmacia, Universidad Complutense, 28040 Madrid, Spain; 3Département de Biologie, Faculté des Sciences, Université Mohammed V-Agdal, Rabat, Morocco
Received for publication December 13, 2004. Accepted for publication September 5, 2005.
ABSTRACT
Genetic variation in 27 populations of Ulex species from southern Spain and northern Morocco (Betic-Rif arc) was assessed using 11 chloroplast microsatellite (cpSSR) markers, which revealed 47 different haplotypes. These nonrecombinant, haploid markers allow measurement of genetic variation in closely related species of Ulex where molecular phylogenetic analyses have not provided a clear view of interspecific relationships. Discriminant analysis indicates that the haplotypes are useful to differentiate among species, and analysis of molecular variance (AMOVA) shows high levels of differentiation among populations. The minimum spanning tree (MST), that represents the connections between the haplotypes, suggests that the eastern Rifean U. africanus haplotypes are more genetically related than those from southern Spain. The latter may have lost genetic diversity while colonizing new habitats, eventually differentiating into U. baeticus and U. scaber. Hybridization between these populations, followed by polyploidization, may have originated the tetraploids (U. congestus and U. borgiae) that colonized new habitats associated with acidic rocks. Separate groupings of U. scaber populations may indicate multiple origins from different stocks. Diversification in this group of Ulex species could be related to the opening of the Alboran Sea by Middle Miocene, when the populations from Morocco and Spain became isolated from each other.
Key Words: chloroplast microsatellites • cpSSRs • genetic variation • Leguminosae • northern Morocco • southern Spain • Ulex
Ulex L. is a small homogeneous genus belonging to the tribe Genisteae (Adans.) Benth., morphologically characterized by a strong reduction of the leaves (often small spiny phyllodes in the mature plant) and a calyx fully divided into two lips. Wild populations of Ulex are restricted to western Europe and northwestern Africa (Morocco and Oran, Algeria), although the greatest diversity is observed in the Iberian Peninsula. Guinea and Webb (1968
, p. 102) stated that "the taxonomy of this genus is very imperfectly understood" and adopted a strongly synthetic approach, recognizing only seven species (with 13 subspecies). For example, many species previously described (e.g., U. eriocladus C. Vicioso, U. australis Clemente, U. jussiaei Webb, and U. baeticus Boiss.) were merged under a broad concept of U. parviflorus Pourret. Based on morphology, chromosome numbers, pollen micromorphology, ecology, and distribution, most of these misinterpreted taxa have since been restored to specific rank (e.g., Cubas, 1984
, 1986
; Rivas Martínez and Cubas, 1987
; Cubas and Pardo, 1992
, 1997
; Pardo et al., 1994
; Espíritu Santo et al., 1997
). As a result of this research, 15 species (and six subspecies) of Ulex are now recognized in Flora Iberica (Cubas, 1999
).
Polyploidy has been an important factor in the evolution of Ulex. Different ploidy levels within species are usually associated with morphological differentiation, separate geographic areas, and/or ecological preferences. Infraspecifically different ploidy levels are recognized taxonomically as subspecies, composing nine diploid, seven tetraploid, and five hexaploid taxa (Cubas, 1986
; Misset, 1990
). For example, U. europaeus subsp. europaeus is hexaploid with a very large area of distribution (from Portugal to Sweden). It is characterized by small bracteoles, whereas subspecies latebracteatus is tetraploid, has very large bracteoles, and is restricted to sandy coastal places along the northwestern Iberian Peninsula (Cubas and Pardo, 1997
). The origin of these (and other) Ulex polyploids is still unresolved, although it is probable that reticulate evolution (through hybridization followed by polyploidization events) has been involved in their origins (Cubas, 1999
).
Despite all these studies, the North African Ulex populations have scarcely been studied. These populations are morphologically close to those growing on the Spanish side of the Gibraltar Strait, but they have a complex pattern of variation that has frustrated the accurate delimitation of taxa and contributed to nomenclatural instability. In addition, the complex geology of the area (IGME, 1980
; Service Geologique du Maroc, 1985
) has resulted in a high diversity of habitats.
This work focuses on the Ulex populations growing along the Spanish Betics and the Moroccan Rif, an arc-shaped mountain belt within the westernmost Mediterranean realm (Fig. 1). These mountains are part of a single orogenic system that formed during the Alpine compressions in Early Miocene time (African–Iberian collision) (e.g., Doblas and Oyarzun, 1989
; Platt and Vissers, 1989
). However, collapse of the mountain belt and subsequent formation of the Alboran Sea by the Middle Miocene led to the formation of a major barrier to gene flow between the populations on either side and may have played an important role in their diversification. In the Betics, three closely related taxa are recognized although their taxonomic status has suffered several rearrangements: (1) U. baeticus, which includes populations on limestones, dolostones, and peridotites (partially serpertinised) in the Ronda sector of the Betics mountains; (2) U. scaber Kunze (syn. U. baeticus subsp. scaber (Kunze) Cubas) grows on marls and calcareous rocks in a restricted area of Cadiz province with a reduced number of individuals isolated due to the intensive land use; and (3) U. borgiae Rivas Martínez [syn. U. scaber var. glabrescens (Webb), U. baeticus subsp. glabrescens (Webb) Cubas], grows on sandstones of the southeastern part of Cadiz province and extends to the border with Malaga province, an area now partly protected under Natural Park regulations. The areas of the three taxa are in contact and even overlap in particular locations (Fig. 1).
|
ascribes the Moroccan populations to (1) U. parviflorus subsp. africanus (Webb) Greuter (syn. U. africanus Webb), widely distributed from Algeria (Oran) to Morocco, and U. scaber (f. genuinus and f. congestus Webb) in the western Rif area. In this work, and until a final taxonomic agreement is attained, we have grouped the Moroccan populations (Fig. 1) into two units we refer to as (1) U. congestus (Webb) Pau, which includes the populations of the northern part of the western Rif area around Tetouan, and (2) U. africanus Webb, comprising the rest of the populations distributed from the western Rif mountains (around Chefchaouen) to Algeria (Oran) and growing on different substrates such as limestones, clays, and marls.
Morphological, serological, and phytogeographic evidence indicated that Ulex represents an extreme in differentiation within the Genista group in the tribe Genisteae (Bisby, 1981
; Feoli-Chiapella and Cristofolini, 1981
). Recent phylogenetic analyses based on nucleotide sequences of nrDNA (ITS region) and cpDNA (trnL-trnF intergenic spacer) confirm that Ulex forms a monophyletic group derived from within the Genista lineage (Aïnouche et al., 2003
; Pardo et al., 2004
). However, compared to other genera of the tribe, Ulex shows weak molecular divergence at the specific level, and molecular phylogenetic studies based on these markers neither clarify the relationships within the genus nor help to establish interspecific relationships within Ulex, particularly for the populations under study (results not shown).
Microsatellites (SSRs, simple sequence repeats) are abundant polymorphic elements of the eukaryotic genome that consist of tandem repetitions of mono-, di-, or tetranucleotide units. These small repeats are prone to expand by particular mechanisms, among which polymerase slippage during replication is probably the most important (Vienne et al., 2003
). Size variation of microsatellites, due to a variable repeat copy number, can be visualized by PCR with pairs of flanking primers and electrophoretic separation of the amplified products (Weising and Gardner, 1999
). Microsatellites typically measure genetic variation for loci that are neutral because the tandem repeats are usually located in noncoding segments of DNA (Frankham et al., 2004
). Chloroplast microsatellites (cpSSRs) that are nonrecombinant, most often uniparentally inherited in angiosperms and effectively haploid, represent potentially useful markers to gain a first insight into the genetic relationships of closely related populations (Provan et al., 1999
, 2001
). Polymorphisms in microsatellite size have been used to detect genetic diversity (Powell et al., 1995
, 1996
) and differentiation and spatial structure among and within populations (Petit et al., 1997
; Walter and Epperson, 2004
). They have proved useful in elucidating the relationships between taxa and in resolving taxonomic uncertainties (Bucci et al., 1998
; Gugerli et al., 2001
; Parducci et al., 2001
). A difficulty in widespread use of cpSSRs as genetic markers is the design of specific primers. However, Weising and Gardner (1999)
produced a set of "universal" primers for amplifying specific chloroplast introns or intergenic spacers, using the fully sequenced tobacco chloroplast genome and other available DNA sequences from different species. These primers revealed polymorphism in species belonging to different families from Solanaceae to Poaceae, including Pisum sativum (Fabaceae).
We have used these universal primers to detect intraspecific cpSSR polymorphisms in closely related Ulex populations from southern Spain and northern Africa, where ITS or chloroplast sequence data are not variable enough to resolve their relationships. This approach has to be regarded as a first step in the discovery of molecular markers that could be used to investigate evolutionary patterns in Ulex and other related genera in the Genisteae tribe. This study aims to (1) investigate the usefulness of cpSSRs for assessing genetic variability at the population level in the genus Ulex, (2) resolve taxonomic uncertainties involving closely related Ulex populations growing around the Rif-Betic arc and separated by the Alboran Sea, and (3) establish the relationships and a plausible origin for the polyploid populations of Ulex in this area.
MATERIALS AND METHODS
Samples
Twenty-five natural populations (207 individuals) of Ulex were analyzed at 11 cpSSR loci. The samples were collected in the field by the authors at different altitudes and on different substrates. The names, provenance, ecology, and voucher number of the samples are indicated in the Appendix and Fig. 1. Based on morphological criteria, the populations were ascribed to five groups: U. baeticus, U. borgiae, U. scaber, U. congestus, and U. africanus. Two additional populations (16 individuals) of U. parviflorus were included in the first assays as an external control of the genetic variability of the group.
Extraction, amplification, sizing, and sequencing
Total DNA was extracted from young or older twigs of 223 plants silica-gel dried using a DNeasy Plant Mini kit (Qiagen, Courtaboeuf, France) following the manufacturer's protocol. Ten chloroplast loci were initially assayed using the consensus primer pairs ccmp1–ccmp10 developed by Weising and Gardner (1999)
. A new cpSSR, 108, within the trnL-trnF intergenic spacer (IGS) region, was amplified using primer e (Taberlet et al., 1991
) as the forward and a reverse primer that was newly designed based on the aligned matrix of a set of sequences of Genisteae species obtained in our previous work (Cubas et al., 2002
; Pardo et al., 2004
; and unpublished data); the sequence of this reverse primer 108 is 5'-ATTCCACTCAGATCCATTTG-3'.
Simple PCR amplifications for individual loci were run initially, then multiplex reactions were performed combining two (ccmp1 and ccmp6) or three primer pairs (ccmp2, ccmp4, and ccmp7; ccmp3, ccmp5, and ccmp10). Microsatellite 108 was amplified separately. PCR amplifications were carried out using a PTC-100 Peltier Thermal Cycler (MJ Research, Basel, Switzerland) in a 20-µL (simple PCR) or 25-µL (multiplex reactions) volume containing 0.2 mmol/L of dNTPs, 0.5–1 unit of DNA polymerase, and 25–50 ng of DNA template. In multiplexed reactions, previous assays showed that equimolar primer concentrations gave uneven amplification results. Following the suggestions of Henegariu et al. (1997)
, several reactions were carried out to find empirically the combination of primer concentrations that gives an even signal for all the loci. The concentration of the primer pairs ranged from 0.4 to 0.7 µmol/L (specific details for each combination are available by request from the authors). Also, a better signal was obtained by raising the concentration of MgCl2 to 3.2 mM. One of the two PCR primers in each pair was 5'-fluorescein-labeled. The PCR reaction was performed with an initial denaturation at 94°C x 5 min, followed by 35 cycles with denaturation at 94°C x 1 min, annealing at 50°C x 1 min, and extension at 72°C x 1 min. A final 15-min extension at 72°C was used to ensure that the terminal transferase reaction by the polymerase was carried out to completion.
Sizing of the amplified products was performed using an automated laser fluorescence DNA sequencer (Applied Biosystem 3730 DNA analyzer; Applied Biosystems, Foster City, California, USA). Fragment sizes expressed in base pairs (bp) were calculated using GeneMapper Software version 3.5 (Applied Biosystems) by comparison with an internal molecular marker standard. Several runs of sizing at different concentrations and from different replications of the PCRs were performed to confirm the fragment length. Because there was no previous information regarding the composition of these loci in Ulex, selected polymorphic amplification products of different sizes were sequenced to detect the presence of the microsatellite region in the amplified fragments. This procedure was also used to confirm that the polymorphism was due to either expansion or contraction of the nucleotide repeat region, and not to mutation in the flanking regions. PCR conditions and purification of the amplicons followed the procedure of Cubas et al. (2002)
. Sequencing was performed on both strands using the BigDye Terminator v3.1 Cycle Sequencing kit (Applied Biosystems) with the amplification primer. Sequencing reactions were electrophoresed on an Applied Biosystem 3730 DNA analyzer. Sizing and sequencing of the amplified products were carried out at the Unidad de Genómica (Parque Científico de Madrid-Universidad Complutense, Madrid, Spain).
Statistical analysis
Because the chloroplast genome does not normally recombine, each unique combination of alleles across the cpSRR loci was scored as a different haplotype. Here the term locus refers to a cpSSR site, and allele refers to a size variant at a given cpSSR site. Because no data are available regarding the best model to explain the evolution of these microsatellites in Ulex, we calculated and compared statistical parameters based on both the infinite allele model (IAM) and the stepwise mutation model (SMM) to analyze the molecular diversity of the samples. The IAM model is based on the number of different alleles, while the SMM model also takes into account differences in allele size.
Haplotype variation within populations was calculated by estimating the number of polymorphic loci (S), the total number of haplotypes (No), the effective number of haplotypes (Ne), and the unbiased haplotype diversity (He) (Nei, 1987
). Genetic distance between individuals within populations was calculated using the IAM model (average gene diversity over loci, 
n; Nei, 1987
), and the SMM model (mean pairwise haplotype distance, D2SH; Vendramin et al., 1998
). A minimum spanning tree (MST) was computed from the matrix of pairwise distances calculated between all pairs of haplotypes using the sum of the squared number of repeat differences between two haplotypes. Discriminant analysis of the haplotypes was carried out with the SAS (1999–2001)
statistical software.
The apportionment of genetic variation within and among populations was determined by analysis of molecular variance (AMOVA; Excoffier et al., 1992
), using ARLEQUIN v. 2.000 (Schneider et al., 2000
). The significance of the fixation indices was tested using a nonparametric permutation approach with 1000 permutations. Distances between haplotypes were calculated from the number of different alleles (IAM model) and the sum of the square size difference in allele size (SMM model) and provided the fixation indices of genetic differentiation among populations FST (overall genetic divergence among populations, obtained with the IAM model) (Weir and Cockerham, 1984
) and RST (overall genetic divergence among populations, obtained with the SMM model) (Slatkin, 1995
), when we estimated genetic structure indices. The degree of relatedness between the genetic distance matrices generated by the two types of indices (IAM and SMM-based) was measured using the Pearson product-moment correlation coefficient.
An unrooted phylogram was constructed by neighbor joining with PAUP* v. 4.0b10 for Macintosh (Swofford, 2002
) based on the matrix of coancestry coefficients for the diploid taxa. The coancestry coefficient (Reynolds et al., 1983
) is the basis of a measure of genetic distance when the divergence between populations with a common ancestral population may be attributed solely to genetic drift.
RESULTS
Morphological and taxonomic considerations
The populations in southern Spain and northern Morocco are morphologically close; however, they do show differences in morphology that permit their separation into different groups. The characters with more variability and the ranges of variation in the studied populations are listed in Table 1. These characters refer to the indumentum of shoots; the size, shape and indumentum of the calyx; and the chromosome number. However, intermediate states are often found and blur the limits of taxa in some cases. In particular, the populations of U. africanus in the westernmost sector of the Rif mountains vary significantly in hair cover and the dimensions of the calyx, as well as in the general indumentum of the plant.
|
; H. Tahiri et al., unpublished data). The tetraploid plants correspond to U. congestus and U. borgiae; they are fertile, produce viable seeds and pollen, and are abundant, forming well-established communities.
Characterization and levels of polymorphism shown by cpSSRs
Nine of the 11 chloroplast loci initially assayed were successfully amplified. Loci ccmp1 and ccmp6 were monomorphic in all the samples. Loci ccmp2, ccmp3, ccmp4, ccmp7, ccmp10, and 108 were polymorphic giving 26 alleles among 223 individuals. Locus ccmp5 amplified successfully but gave a multiband pattern and so was excluded from the analysis. Loci ccmp8 and ccmp9 failed in all the assays and, consequently, were not used. The presence and composition of microsatellites were characterized and confirmed by sequencing selected polymorphic amplification products. Table 2 shows the size of the amplified fragments and the number and composition of the microsatellites. The best six polymorphic loci were used for the genetic analysis, i.e., ccmp2, ccmp3, ccmp4, ccmp7, ccmp10, and 108.
|
0.035, eight or more individuals). There are marked differences in the distribution of haplotypes within each taxon. The seven U. africanus populations have 13 haplotypes; four of them (h26, h27, h32, and h34) are shared by two or three populations. On the contrary, the eight haplotypes in the three U. scaber populations are private for a population. Similarly, all 13 haplotypes in the six U. baeticus populations are population-specific. In the seven populations of U. borgiae, there are 12 haplotypes; only one is shared by two populations. In the case of U. congestus, one population (UCGG) has one haplotype, while the other (UCDE) is much more variable with three.
|
Table 4 shows the estimates of chloroplast variation based on the six cpSSR loci in the 27 populations studied. Haplotype diversity (He) has a high range of variation in all the taxa: 0– 0.68 in U. baeticus, 0.20–0.82 in U. scaber, 0–0.54 in U. borgiae, 0–0.60 in U. congestus, and 0.25–0.81 in U. africanus. The two populations of U. parviflorus are fixed (UPTO and UPVE; He = 0).
|
|
Discriminant analysis of the haplotype data, using the diploid taxa (U. africanus, U. scaber, and U. baeticus) as the classification variable, separates into differentiated groups most of the haplotypes found in each of the taxa (Fig. 3). As expected, the haplotypes shared between taxa (i.e., h12 and h39) lay in overlapping areas. In addition, the position of several haplotypes indicates the relationships between different taxa. Three haplotypes of U. africanus (h29 and h28, UMBN; and h40, in one of eight individuals of UMTT) are close to those of the U. baeticus and, similarly, one haplotype of U. scaber (h30, found in two of eight individuals of USSJ). Conversely, one U. baeticus haplotype (h1, present in six of eight individuals of UBBE) is closer to the U. africanus haplotypes. The analysis classified correctly 87.76% of the individuals of U. baeticus, 61.54% of U. scaber individuals, and 81.82% of U. africanus (on average 80% of the individuals were correctly classified). Discriminant analysis using all the taxa (not shown) does not group the haplotypes of each taxon, except for a tendency to separate most of those of U. africanus. The haplotypes of the U. congestus and U. borgiae populations are intermixed with those of the other taxa. A posteriori classification, based on the discriminant function developed from the first analysis (diploid taxa), ascribed the U. congestus (UCDE and UCGG) and most of the U. borgiae populations (UGCA, UGCO, UGES, and UGGA) to U. baeticus. The rest of the U. borgiae populations (UGJU, UGMO, and UGPI) are classified as U. scaber.
|
n, Table 4) are in U. baeticus populations (UBYU, UBPA, UBSE, and UBBE), while the highest values of D2SH are found in U. baeticus (UBYU, UBSE, and UBBE) and U. scaber (USSJ). However, both estimates are moderately well correlated (r = 0.812185, R2 = 65.96%, P < 0.01). The degree of genetic differentiation among populations was estimated using FST and RST indices (Table 4). Both statistics are also moderately well correlated (r = 0.849263, R2 = 72.12%, P < 0.01). Because no significant difference was found, we assume a stepwise mutation model at the microsatellite loci, considering that the RST statistic provides less-biased estimates of the demographic parameters for a population (Slatkin, 1995
). AMOVA analysis, using as distance the RST index, shows slight differences in the apportionment of genetic diversity within taxa (Table 5). Most of the variation detected is due to differences among populations: 94.93% in U. congestus, 89.72% in U. borgiae and 88.91% in U. scaber, while in U. africanus and U. baeticus this accounts for only 77.88% and 67.13%, respectively. The AMOVA analysis of the diploid taxa (U. scaber, U. baeticus, and U. africanus) shows that 22.02% of the variation is due to differences between taxa and that 55.50% is attributable to differences among populations within taxa. In the case of U. borgiae and U. congestus (the tetraploid taxa), the apportionment of the genetic diversity is different: most of the variation (89.3%) is attributable to differences among populations within taxa and only 1.85% to differences between taxa. These results suggest that U. borgiae and U. congestus are closely related. When AMOVA analysis is conducted for all the populations, differences between taxa account for 20.30% of the variation, while the principal component is due to differences among populations within taxa (64.62%). There is no significant geographic apportionment of genetic diversity between the Moroccan and Spanish populations in the AMOVA analysis when all taxa are included (Table 5). Most of the variation is explained by differences among populations (67.3%), while 18.5% is attributable to the geographic origin of the samples.
|
|
Based on morphological characters, several lines of differentiation have been suggested within the genus Ulex, (e.g., Rothmaler, 1942
; Cubas, 1984
, 1999
). However, the combination of the morphological characters, chromosome numbers, and the topology of the phylogenetic trees suggest reticulate relationships between species. The detection of taxa involved in the origin of polyploids in Ulex is difficult because the levels of molecular divergence between diploid species (populations and individuals) are very small (Aïnouche et al., 2003
; P. Cubas, unpublished data). Although the ITS region nrDNA and cpDNA provide little discriminatory information, the studied cpSSRs in Ulex have proved of value in the detection of within-population genetic variation. In most populations, two to four haplotypes have been detected, and only three polyploid and one diploid populations are monomorphic for all loci. Less intraspecific variation has been found in other angiosperms, for example, among Nicotiana tabacum and Actinidia sinensis cultivars (Weising and Gardner, 1999
), and only three haplotypes were found in 25 wild populations in Prunus spinosa using a similar set of cpSSRs (Mohanty et al., 2002
). On the contrary, in gymnosperms, cpSSRs detect an average of 4–7 haplotypes per population in Pinus halepensis, P. brutia, and P. eldarica (Bucci et al., 1998
), 4–10 haplotypes in P. canariensis (Gómez et al., 2003
), and 11–22 haplotypes per population in Abies species (Parducci et al., 2001
). In contrast to the case in angiosperms, the chloroplast genome in gymnosperms is paternally inherited via pollen (generally with long-distance dispersal by wind), and the primers were designed based on the complete chloroplast sequence of P. thunbergii.
Microsatellites reveal a high level of variation attributable to differences among populations in all the Ulex taxa studied (FST ranging from 0.61 to 0.84), most markedly in U. congestus, U. borgiae, and U. scaber populations. The AMOVA analysis shows that around 90% of the variation is explained by differences among populations. Historical factors, together with limited seed dispersal, may have contributed to the generation of this large genetic structure among populations. Comparatively, the amount of variation attributable to differences between taxa is lower, especially if the tetraploid taxa are included. The AMOVA analysis indicates that 20% of the variation can be explained by differences between taxa. Nevertheless, the discriminant analysis of the haplotypes, using the diploid taxa as the classification factor, indicates that most of the haplotypes are characteristic of a taxon and that they are useful in their differentiation (Fig. 3).
Our study reveals different situations in the diploid taxa (Figs. 2a and 4). (1) In Ulex africanus several haplotypes are shared among populations, most of them interconnected in the MST of the haplotypes; Moreover, most of the populations are close in the NJ coancestry phylogram. (2) In U. baeticus none of the haplotypes are shared by the different populations although most of them are still linked in the MST, and in the NJ tree the populations are partially connected and also linked to population UMBN (included here in U. africanus). (3) In U. scaber the haplotypes are not shared by the populations nor connected as a whole in the MST and are clearly separated in the NJ tree. These data suggest that the populations of U. africanus (at least the eastern ones) still have a certain amount of gene flow via seeds. The data on populations of U. baeticus reflect past relationships, although today they are much more genetically isolated. However, in the case of populations grouped under U. scaber, they either originated from different stocks or have been isolated for so long that the relationships are no longer reflected in the microsatellite variation. Interestingly, these populations are reduced in the number of individuals and are also geographically isolated. The Ulex individuals from the center of the Moroccan area (Targuist area, UMBN) deserve a more critical morphological study, and most probably they should be ascribed to a taxon related to U. baeticus.
The microsatellite analysis also provides useful information on the polyploid taxa (Fig. 2b): (1) U. borgiae and U. congestus share one haplotype, and the other haplotypes are connected in the MST tree; (2) in the AMOVA analysis, little of the variation is attributable to difference between the taxa; and (3) the discriminant analysis and the MST (including diploid and polyploid taxa) position part of the haplotypes of U. borgiae and U. congestus as linked to U. baeticus, whereas others connect to U. scaber haplotypes. The contribution of the U. africanus populations was either minimal or undetectable. Together with the fact that the cpSSR markers used are uniparentally inherited, these data suggest that populations related to U. baeticus and U. scaber were probably involved in the origin of both polyploids and that U. borgiae and U. congestus should most probably be treated as conspecific. The results also suggest that hybridization could have occurred several times between different populations. However, cpSSR markers may not reveal the entire history of divergence and hybridization among these populations, which could involve a significant amount of nuclear introgression. The Ulex populations may have diverged from one another by genetic drift, while gene flow between populations (including hybridization processes) could have broken down this differentiation. Given that organelle genomes are haploid, their effective population size in hermaphrodite outcrossing plants such as Ulex is half that of the diploid nuclear genomes. As a result, chloroplast-specific markers would be better indicators of historical bottlenecks, founder effects, and genetic drift than nuclear markers (Provan et al., 2001
). Nuclear markers are dispersed both by pollen and seed flow, while organellar markers are usually only dispersed by seed flow. In consequence, the genetic diversity of nuclear markers among populations (FST) should be lower than for maternally inherited organelle markers. Thus, the extent of the discrepancy between these two measures of FST depends on the relative rates of pollen and seed flow among population (Ennos et al., 1999
). The joint application of cytoplasmic and nuclear markers would be crucial for disentangling the relative contributions of seed and pollen dispersal to overall levels of interpopulation and interspecific gene flow in this group, and help to definitively settle the origin and taxonomic status of the polyploid taxa.
A plausible evolutionary scenario for Ulex populations in Spain and Morocco
There is no agreement regarding the relationships and taxonomic status of the Ulex populations of southern Spain and northern Morocco along the Rif-Betic arc. These populations are morphologically related and a clear-cut separation between them cannot be drawn. Nevertheless, they show ecological preference trends, and they have different ploidy levels.
In southern Spain (Fig. 1), the diploid U. baeticus grows on limestones, dolostones, and peridotites and constitutes an important component of the shrub vegetation in the Ronda sector (Malaga) of the Betics Mountains. Ulex scaber, also a diploid, grows preferentially on marls and carbonate rocks, although the populations are more restricted and isolated from the intense use of the land. Ulex borgiae grows on sandstones and other acidic substrata in the southern and eastern parts of Cadiz province; the populations are tetraploid in the cases studied, except for one that is hexaploid. However, the topography and geology of this area is complex. In fact, tectonic imbrication of geologic units results in a complex array of remarkably different lithologies within relatively small areas. Similarly, many morphologically intermediate plants are found along the contact zones between geologic units.
The Moroccan case is very similar, although the populations are distributed between two separate areas (Fig. 1). The western sector includes the Tanger area (sandstones) and the western Rif mountains with acidic rock and dolostones (around Tetouan) and limestones (around Chefchaouen). Ulex congestus grows in the Tanger area and around Tetouan on sandstones and dolostones. The plants are morphologically close to U. borgiae, and the available chromosome numbers indicate that they are tetraploid (H. Tahiri et al., unpublished data), although it would be advisable to check this with more samples. In the Chefchaouen area, the plants are diploid and morphologically closer to U. africanus, although slightly differentiated from the eastern populations. The central-eastern zone is dominated by limestones, clays, and marls and ranges in altitude from sea level to 1500 m a.s.l. This area extends from the Central Rif Mountains (around Targuist) to a coastal band (eastern to Melilla and Nador), and penetrates inland up to the mountains of Beni Snassen (near Berkane). The morphology of U. africanus in this area has a wide range of variation, although there is no clear pattern related to geographical position or ecological conditions. The Ulex populations are now scattered as a result of extensive human use of the territory around Targuist, whereas in the eastern sector the populations are more commonly contiguous and better preserved.
The genetic variation detected in this study suggests a plausible scenario for the differentiation of Ulex species in the region. As indicated, the Betics and Rif mountains were part of a single orogenic system, that formed during the Alpine compressions in the Early Miocene (African-Iberian collision). However, by Middle Miocene time, the orogenic edifice underwent extensional collapse, which ultimately led to formation of the Alboran basin (Middle to Upper Miocene; e.g., Doblas and Oyarzun, 1989
; Platt and Vissers, 1989
), separating North Africa from Spain. Thus, the populations of Ulex that extend along the Betic-Rif arc have remained isolated from each other since this time. However, the eastern Rifean U. africanus populations remained genetically connected, whereas the populations in westernmost Rif (Morocco) and southern Spain colonized new habitats associated with dolostones and peridotites, probably differentiating into U. baeticus. Loss of genetic diversity as a result of genetic drift could have been responsible for the origin of the populations of U. scaber. Intensive land use may also have contributed to an increase in the isolation factors affecting both the populations in the western Rif and the U. scaber populations in Cadiz. On the other hand, hybridization between these still genetically related diploid populations, followed by polyploidization, conceivably led to the origin of the tetraploid populations (i.e., U. congestus-borgiae) that in turn colonized new habitats associated with acidic rocks.
A final conclusion is that the analysis of cpSSR molecular markers has proved to be a useful tool to investigate the genetic variation in this closely related group of species of Ulex, for which the rate of morphological diversification has not closely matched that of genetic and ecological differentiation. These markers have revealed interspecific and interpopulation diversification resulting in testable hypotheses regarding the diversification of the group and the origin of the polyploid populations. Widespread cross-amplification with cpSSR markers has been found in Glycine and even between different genera (Peakall et al., 1998
); thus the next step to further elucidate relationships would be to extend the analysis of chloroplast microsatellites to other groups of Ulex. A combination of these and other independent molecular markers can be used to further test the hypothesis proposed here and provide new insight about other Ulex species in which the evolution may also have been reticulate.
|
1 The authors thank D. Hawksworth (Madrid, Spain) for reviewing the manuscript and providing constructive suggestions, M. Garcia (Unidad de Genómica, Parque Científico de Madrid-Universidad Complutense) for invaluable technical assistance, and two anonymous reviewers for helpful comments and suggestions. Funding for this work was provided by the Ministry of Education and Science of Spain (Project REN2002-00225) and the Spanish Agency of International Cooperation (grant to H.T., MAE-AECI, programa II.A, 2003/ 04). 
4 Author for correspondence (cubas{at}farm.ucm.es
)
LITERATURE CITED
Aïnouche A. R. J. Bayer P. Cubas M.-TH. Misset 2003 Phylogenetic relationships within tribe Genisteae (Papilionoideae) with special reference to genus Ulex. In B. Klitgaard and A. Bruneau [eds.], Advances in legume systematics, part 10, 239–252. Royal Botanic Gardens, Kew, UK
Bisby F. A. 1981 Genisteae (Adans.) Benth. In R. M. Polhill and P. H. Raven [eds.], Advances in legume systematics, part 1, 409–425. Royal Botanic Gardens, Kew, UK
Bucci G. M. Anzidei A. Madaghiele G. G. Vendramin 1998 Detection of haplotypic variation and natural hybridization in halepensis-complex pine species using chloroplast simple sequence repeat (SSR) markers. Molecular Ecology 7: 1633-1643[CrossRef]
Cubas P. 1984 Estudio taxonómico de los géneros Ulex L. y Stauracanthus Link en la Península Ibérica. Colección tesis doctorales, 211/84 ed. Universidad Complutense, Madrid, Spain
Cubas P. 1986 Números cromosomáticos en Ulex L. y Stauracanthus Link (Genisteae-Papilionaceae). Anales Jardín Botánico de Madrid 43: 217-233
Cubas P. 1999 Ulex L. In S. Talavera, C. Aedo, S. Castroviejo, C. Romero Zarco, L. Sáez, F. J. Salgueiro, and M. Velayos [eds.], Flora iberica, vol. 2, 212–239. CSIC, Madrid, Spain
Cubas P. C. Pardo 1992 Pollen wall stratification in Ulex (Genisteae, Papilionoideae: Leguminosae) in the Iberian Peninsula. Grana 31: 177-185
Cubas P. C. Pardo 1997 Correlations between chromosomal and morphological characters in subspecies of Ulex europaeus L. (Genisteae, Leguminosae) from the NW of the Iberian Peninsula. Botanical Journal of the Linnean Society 125: 229-243[CrossRef]
Cubas P. C. Pardo H. Tahiri 2002 Molecular approach to the phylogeny and systematics of Cytisus (Leguminosae) and related genera based on nucleotide sequence of nrDNA (ITS region) and cpDNA (trnL-trnF intergenic spacer). Plant Systematics and Evolution 233: 223-242[CrossRef][ISI][Medline]
Doblas M. R. Oyarzun 1989 Neogene extensional collapse in the western Mediterranean (Betic-Rif Alpine orogenic belt): implications for the genesis of the Gibraltar Arc and magmatic activity. Geology 17: 430-433
Ennos R. A. W. T. Sinclair X.-S. Hu A. Langdon 1999 Using organelle markers to elucidate the history, ecology and evolution of plant populations. In P. M. Hollingsworth, R. M. Bateman, and R. J. Gornall [eds.], Molecular systematics and plant evolution, 1–19. Taylor & Francis, New York, New York, USA
Espirito Santo M. D. P. Cubas M. F. Lousa C. Pardo J. C. Costa 1997 Ulex parviflorus sensu lato (Genisteae, Leguminosae) en la zona Centro de Portugal. Anales Jardín Botánico de Madrid 55: 49-65
Excoffier L. P. E. Smouse J. M. Quattro 1992 Analysis of molecular variance inferred from metric distances among DNA haplotypes: application to human mitochondrial DNA restriction data. Genetics 131: 479-491[Abstract]
Feoli-Chiapella L. G. Cristofolini 1981 Serological contributions to the systematics of Ulex (Genisteae-Fabaceae) and allied genera. Nordic Journal of Botany 1: 723-729
Frankham R. J. D. Ballou D. A. Briscoe 2004 A primer of conservation genetics. Cambridge University Press, Cambridge, UK
Gomez A. S. C. Gonzalez-Martinez C. Collada J. Climent L. Gil 2003 Complex population genetic structure in the endemic Canary Island pine revealed using chloroplast microsatellite markers. Theoretical and Applied Genetics 107: 1123-1131[CrossRef][ISI][Medline]
Gugerli F. J. Senn M. Anzidei A. Madaghiele U. Buchler C. Sperisen G. G. Vendramin 2001 Chloroplast microsatellites and mitochondrial nad1 intron2 sequences indicate congruent phylogenetic relationships among Swiss stone pine (Pinus cembra), Siberian stone pine (Pinus sibirica), and Siberian dwarf pine (Pinus pumila). Molecular Ecology 10: 1489-1497[CrossRef][Medline]
Guinea E. D. A. Webb 1968 Ulex L. In T. G. Tutin, V. H. Heywood, N. A. Burges, D. M. Moore, D. H. Valentine, S. M. Walters, and D. A. Webb [eds.], Flora Europaea, vol. 2, 102–103. Cambridge University Press, UK
Henegariu O. N. A. Heerema S. R. Dlouhy G. H. Vance P. H. Vogt 1997 Multiplex PCR: critical parameters and step-by-step protocol. BioTechniques 23: 504-511[ISI][Medline]
IGME. 1980 Mapa geológico de la Península Ibérica, Baleares y Canarias (escala 1:1000000). Ministerio de Industria y Energía, Servicio de Publicaciones, Madrid, Spain
Maire R. 1987 Flore de l'Afrique du Nord, vol. 16, editions Lechevalier. S.A.R.L., Paris, France
Misset M.-TH. 1990 Données caryologiques chez le genre Ulex L. (Papilionoideae) dans le Massif Armoricaine. Taxon 39: 630-635[CrossRef][ISI]
Mohanty A. J. P. Martin I. Aguinagalde 2002 Population genetic analysis of European Prunus spinosa (Rosaceae) using chloroplast DNA markers. American Journal of Botany 89: 1223-1228
Nei M. 1987 Molecular evolutionary genetics. Columbia University Press, New York, New York, USA
Pardo C. P. Cubas P. Sanchez Testillano 1994 Tendances evolutives de la structure exinique dans les genres Ulex, Stauracanthus et Genista (Genisteae, Papilionoideae: Leguminosae). Acta Botanica Gallica 141: 195-205[ISI]
Pardo C. P. Cubas H. Tahiri 2004 Molecular phylogeny and systematics of Genista (Leguminosae) and related genera based on nucleotide sequences of nrDNA (ITS region) and cpDNA (trnL-trnF intergenic spacer). Plant Systematics and Evolution 244: 93-119[CrossRef][ISI]
Parducci L. A. E. Szmidt A. Madaghiele M. Anzidei G. G. Vendramin 2001 Genetic variation at chloroplast microsatellites (cpSSRs) in Abies nebrodensis (Lojac.) Mattei and three neighboring Abies species. Theoretical Applied Genetics 102: 733-740[CrossRef]
Peakall R. S. Gilmore W. Keys M. Morgante A. Rafalsky 1998 Cross-species amplification of soybean (Glycine max) simple sequence repeats (SSRs) within the genus and other legume genera: implications for the transferability of SSRs in plants. Molecular Biology and Evolution 15: 1275-1287[Abstract]
Petit R. J. E. Pineau B. Demeseure R. Bacilieri A. Ducousso A. Kremer 1997 Chloroplast DNA footprints of postglacial recolonization by oaks. Proceedings of the National Academy of Sciences, USA 94: 9996-10001
Platt J. P. R. L. M. Vissers 1989 Extensional collapse of thickened continental lithosphere: a working hypothesis for the Alboran Sea and Gibraltar arc. Geology 17: 540-543
Powell W. M. Morgante R. McDevitt G. G. Vendramin J. A. Rafaslki 1995 Polymorphic simple sequence repeat regions in chloroplast genomes: applications to the population genetics of pines. Proceedings of the National Academy of Sciences, USA 92: 7759-7763
Powell W. G. Machray J. Provan 1996 Polymorphism revealed by simple sequence repeats. Trends in Plant Science 1: 215-222
Provan J. N. Soranzo N. J. Wilson J. W. McNicol M. Morgante W. Powell 1999 The use of uniparentally inherited simple sequence repeat markers in plant population studies and systematics. In P. M. Hollingsworth, R. M. Bateman, and R. J. Gornall [eds.], Molecular systematics and plant evolution, 35–50. Taylor & Francis, New York, New York, USA
Provan J. W. Powell P. M. Hollingsworth 2001 Chloroplast microsatellites: new tools for studies in plant ecology and evolution. Trends in Ecology and Evolution 16: 142-147
Reynolds J. B. S. Weir C. C. Cockerham 1983 Estimation of the coancestry coefficient: basis for a short-term genetic distance. Genetics 105: 767-779
Rivas Martinez S. P. Cubas 1987 Ulex L. In B. Valdés, S. Talavera, and E. Fernández-Galiano [eds.], Flora vascular de Andalucía Occidental, 2: 165–169. Ketres Editora, Barcelona, Spain
Rothmaler W. 1942 Revision der Genisteen. I. Monographien der gattungen um Ulex. Botanische Jahrbücher für Systematik 72: 69-116
SAS. 1999–2001 SAS system for Windows, release 8.02. SAS Institute, Cary, North Carolina, USA
Schneider S. D. Roessli L. Excoffier 2000 Arlequin version 2.000: a software for population genetics data analysis. Genetics and Biometry Laboratory, University of Geneva, Basel, Switzerland
Service Geologique du Maroc. 1985 Carte géologique du Maroc (échelle 1:1000000). Notes et mémoires du Service Géologique, no. 260, Rabat, Morocco
Slatkin M. 1995 A measure of population subdivision based on microsatellite allele frequencies. Genetics 139: 457-462[ISI][Medline]
Swofford D. L. 2002 PAUP*: phylogenetic analysis using parsimony (*and other methods), version 4.0b10. Sinauer, Sunderland, Massachusetts, USA
Taberlet P. L. Gielly G. Pautou J. Bouvet 1991 Universal primers for amplification of three non-coding regions of chloroplast DNA. Plant Molecular Biology 17: 1105-1109[CrossRef][ISI][Medline]
Vendramin G. G. M. Anzidei A. Madaghiele G. Bucci 1998 Distribution of genetic diversity in Pinus pinaster Ait. as revealed by chloroplast microsatellites. Theoretical and Applied Genetics 97: 456-463[CrossRef][ISI]
Vienne D. de S. Santoni M. Falque 2003 Principal sources of molecular markers. In D. de Vienne [ed.], Molecular markers in plant genetics and biotechnology, 1–46. Science Publishers, Plymouth, UK
Walter R. B. K. Epperson 2004 Microsatellite analyses of spatial structure among seedlings in populations of Pinus strobus (Pinaceae). American Journal of Botany 91: 549-557
Weising K. R. C. Gardner 1999 A set of conserved PCR primers for the analysis of simple sequence repeat polymorphisms in chloroplast genomes of dicotyledonous angiosperms. Genome 42: 9-19[Medline]
Weir B. S. C. C. Cockerham 1984 Estimating F-statistics for the analysis of population structure. Evolution 38: 1358-1370[CrossRef][ISI]
This article has been cited by other articles:
|
|
 |
|
  C. Pardo, P. Cubas, and H. Tahiri Genetic variation and phylogeography of Stauracanthus (Fabaceae, Genisteae) from the Iberian Peninsula and northern Morocco assessed by chloroplast microsatellite (cpSSR) markers Am. J. Botany, January 1, 2008; 95(1): 98 - 109. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
